Over the past few years, we completed human trials of recombinant glycoprotein vaccines for treatment and prevention of genital herpes. The vaccines proved inadequate. We turned, therefore, to animal model studies using new classes of HSV vaccines to define ones that appear to be more immunogenic and potentially more effective. We are comparing recombinant glycoprotein vaccines with novel DNA-based and live, genetically-engineered vaccines for genital herpes. This has the theoretical advantage of inducing better CTL responses. The DNA vaccine expresses gD2 under a strong CMV promoter. The engineered virus vaccine termed dl5/29 was made by David Knipe at Harvard and consists of HSV-2 deleted for genes 5 and 29, impairing replication and establishment of latency. In mice and guinea pigs dl5/29 is as prtective as recombinant gD2 against acute disease, latency and recurrence rates. In guniea pigs, dl5/29 is more immunotherapeutic than recombinant protein in CFA/IFA. Immunogenicity studies show strong antibody and cellular responses elicited by by dl5/29. Further studies will be quantifying the specific components of immune repsonses elicited by each of these candidate vaccines.